Anti‑HER3 monoclonal antibody exerts antitumor activity in a mouse model of colorectal adenocarcinoma.

HER3 belongs to the epidermal growth factor receptor (EGFR) family and is known to form an active heterodimer with other three family members EGFR, HER2, and HER4. HER3 is overexpressed in lung, breast, colon, prostate, and gastric cancers. In the present study, we developed and validated an anti‑HER3 monoclonal antibody (mAb), H3Mab‑17 (IgG2a, kappa), by immunizing mice with HER3‑overexpressed CHO‑K1 cells (CHO/HER3). H3Mab‑17 was found to react specifically with endogenous HER3 in colorectal carcinoma cell lines, using flow cytometry. The KD for H3Mab‑17 in CHO/HER3 and Caco‑2 (a colon cancer cell line) were determined to be 3.0x10‑9 M and 1.5x10‑9 M via flow cytometry, respectively, suggesting high binding affinity of H3Mab‑17 to HER3. Then, we assessed the H3Mab‑17 antibody‑dependent cellular cytotoxicity (ADCC) and complement‑dependent cytotoxicity (CDC) against Caco‑2, and evaluated its antitumor capacity in a Caco‑2 xenograft model. In vitro experiments revealed H3Mab‑17 had strongly induced both ADCC and CDC against Caco‑2 cells. In vivo experiments on Caco‑2 xenografts revealed that H3Mab‑17 treatment significantly reduced tumor growth compared with the control mouse IgG. These data indicated that H3Mab‑17 could be a promising treatment option for HER3‑expressing colon cancers.

Introduction

The epidermal growth factor receptor (EGFR) family, also known as HER or ErbB, has a tyrosine kinase domain in its intracellular region (1). The EGFR family transduces extracellular to intracellular signals through the activation of tyrosine kinase domain (1). By binding to the ligand, the extracellular domain promotes the formation of homodimers or heterodimers between the EGFR family receptors (2,3). This dimerization is essential for the activation of the tyrosine kinase domain and intracellular signaling pathways such as Ras/MAPK, PI3K/Akt, and JAK/STAT (4,5).

The EGFR family consists of four members [EGFR (HER1, ErbB1), HER2 (ErbB2), HER3 (ErbB3), and HER4 (ErbB4)] and each member has different ligands: EGFR binds to seven ligands such as EGF, TGF-α, and epigen; HER3 binds to neuregulin1 and neuregulin2; HER4 binds to seven ligands such as heparin binding-EGF, betacellulin, and epiregulin. In contrast, there is no ligand for HER2 (6,7). Although HER3 has a tyrosine kinase domain, its kinase activity is impaired (8,9). Therefore, transphosphorylation by other members of the EGFR family is required to activate HER3. HER3 can form an active heterodimer with the other three members of the EGFR family (2,10–13).

The EGFR family plays an essential role in regulating cell growth and in the differentiation, proliferation, and survival of normal cells. Insufficient EGFR signaling is associated with Alzheimer's disease and multiple sclerosis (14), while the overexpression of EGFR family is associated with the development of tumors (15–17). The EGFR family has been found to be overexpressed in many cancers as below: EGFR in breast, non-small cell lung, and prostate cancers (18); HER2 in breast, colon, lung, and pancreatic cancers (18); HER3 in lung, breast, colon, prostate, and stomach cancers (2,19); HER4 in non-small cell lung, and ovarian cancers (20,21). Therefore, the EGFR family is thought to be a valid target for candidates in cancer therapy.

High expression of HER3 is thought to be an established negative prognostic factor in several solid tumors including colorectal cancer (22,23). Metastatic colorectal cancer is one of the most aggressive tumors, associated with high mortality rates worldwide (24). In a previous study, 79% of primary tumors were found to present a high HER3 expression and there was a correlation between HER3 expression in primary tumors and corresponding lymph node metastases in 236 colorectal cancer patients (25). In addition, elevated HER3 expression was associated with shorter overall survival and disease-free survival in patients with colorectal cancer (25). Furthermore, HER3 downregulation in colorectal cancer cell lines caused G2-M cell-cycle arrest, leading to apoptosis and abrogated cell proliferation, migration, and invasion (22). Altogether, these results suggest that HER3 can be a potential therapeutic target for colorectal cancer.

Several monoclonal antibodies (mAbs) have been established as an innovative immunotherapy against tumors. Programmed cell death 1 (PD-1) and cytotoxic T-lymphocyte-associated protein 4 (CTLA-4) are inhibitory receptors for immune checkpoints, which are expressed on the surface of T cells (26–29). Anti-PD-1 and anti-CTLA-4 mAbs have been reported as potential anticancer drugs (30,31). Nivolumab and pembrolizumab are anti-PD-1 mAbs, which were approved by the US Food and Drug Administration (32–34). Both mAbs activate the immune system to attack tumors, by blocking the interaction between PD-1 and its ligand, PD-L1, which is expressed in cancer cells. DNA mismatch repair deficient tumors have very high levels of DNA microsatellite instability, and it is known that microsatellite instable tumors have highly upregulated expression of multiple immune checkpoint proteins, including PD-1 compared with microsatellite stable tumors; therefore, nivolumab and pembrolizumab are available for the treatment of DNA mismatch repair deficiency and microsatellite instable subset of colorectal cancer (35–37).

Several antibody drugs have been developed against ligands, such as transforming growth factor (TGF)-α and EGF, or receptors, such as EGFR (38). These mAbs neutralize the interaction between ligands and receptors. Antibody-drug conjugate (ADC) is a complex molecule, which is composed of an antibody, linker, and an anticancer drug, and delivers the anticancer drug to target cells (39). Moreover, some mAbs possess antibody-dependent cellular cytotoxicity (ADCC) or complement-dependent cytotoxicity (CDC). Cetuximab, a mouse/human chimeric IgG1 against EGFR, binds to the ligand-binding site of EGFR, and inhibits the activation and dimerization of EGFR (38). Cetuximab has been used for the treatment of metastatic colorectal cancer, metastatic non-small cell lung cancer, and head and neck squamous cell carcinomas (HNSCC) (38). Trastuzumab, a humanized mAb against HER2, has been used to treat HER2-positive cancers, such as breast cancers and gastric cancers (40,41). Trastuzumab binds to the extracellular domain of HER2 and downregulates activation of AKT (42). Moreover, trastuzumab exhibited ADCC in a mouse model (43). However, it has been shown that some types of cancers are resistant to cetuximab and trastuzumab (44,45).

Most HNSCCs are resistant to cetuximab, because cetuximab treatment induces HER2/HER3 dimerization and HER3 activation in HNSCC cell lines (44). It has been reported that anti-HER3-ADC exerts antitumor effect on breast cancer cells, which have resistance to trastuzumab (45). For this reason, the development of anti-HER3 mAbs has been required for cancer therapy. Seribantumab and lumretuzumab are anti-HER3 mAbs, which block HER3-neuregulin interaction and inhibit HER3 heterodimerization and phosphorylation (46,47). Lumretuzumab is also known to have ADCC activity (46). Phase II and phase Ib/II trial are now ongoing concerning seribantumab and lumretuzumab, respectively (47,48). An anti-HER3-ADC (U3-1402), composed of an anti-HER3 mAb (patritumab) and a novel topoisomerase I inhibitor (DX-8951 derivative; DXd) has entered phase I and II trials for the treatment of HER3-positive non-small cell lung cancers (NCT04676477), metastatic breast cancers (NCT02980341), and colorectal cancers (NCT04479436) (49–51). Preliminary results demonstrate that U3-1402 treatment appears to be safe and exhibits antitumor activity, suggesting that HER3-targeting therapy may be effective for HER3-overexpressing metastatic breast cancers (50).

It has been reported that one amino acid substitution in EGFR in tumors causes acquisition of resistance to gefitinib after gefitinib treatment (52,53); therefore, HER3 may also acquire resistance to seribantumab and lumretuzumab in the future. To characterize the HER3 and HER3-targeting cancer therapy, the development of further anti-HER3 specific mAbs is required. In this study, we developed a novel anti-HER3 mAb against colon cancers using a Cell-Based Immunization and Screening (CBIS) method (54). Furthermore, we investigated whether a novel anti-HER3 mAb shows ADCC/CDC activities or antitumor activities for colon cancers.

Materials and methods

Construction of plasmids

The Genome Network Project clone IRAK174J18 (HER3) was provided by the RIKEN BioResource Research Center through the National BioResource Project of the MEXT and AMED agencies of Japan. HER3 DNA plus N-terminal PA16 tag, recognized by NZ-1, was subcloned into a pCAG-Ble vector (FUJIFILM Wako Pure Chemical Corp.) and named pCAG/PA16-HER3. HER3 DNA plus C-terminal PA tag, recognized by NZ-1, was subcloned into a pCAG-Neo vector (FUJIFILM Wako Pure Chemical Corp.) and named pCAG/HER3-PA.

Cell lines

A mouse myeloma cell line (P3X63Ag8U.1; P3U1), Chinese hamster ovary (CHO)-K1 cells, a glioblastoma cell line (LN229), colorectal adenocarcinoma cell lines (Caco-2, LS 174T, COLO 201, HCT-8, SW1116, and HT-29), and a colorectal carcinoma cell line (HCT 116) were obtained from the American Type Culture Collection. Colon adenocarcinoma cell lines (HCT-15, COLO 205, and DLD-1) and a breast adenocarcinoma cell line (MCF7) were obtained from the Cell Resource Center for Biomedical Research Institute of Development, Aging and Cancer at Tohoku University. CHO/PA16-HER3 and CHO/HER3-PA were established by transfecting pCAG/PA16-HER3 and pCAG/HER3-PA, respectively, into CHO-K1 cells using the Neon Transfection System (Thermo Fisher Scientific, Inc.). A few days after transfection, cells positive for anti-HER3 mAb (clone D22C5; cat. no. 12708; Cell Signaling Technology, Inc.) were sorted using a cell sorter (SH800; Sony Biotechnology Corp.). CHO/mock (Ble) and CHO/mock (Neo) were established by transfection of the pCAG-Ble vector and pCAG-Neo vector, respectively. Stable transfectants of CHO/mock (Ble) and CHO/PA16-HER3 were cultured at 37°C for 14 days on media containing 0.5 mg/ml of Zeocin (InvivoGen), and stable transfectants of CHO/mock (Neo) and CHO/HER3-PA were cultured at 37°C for 14 days on media containing 0.5 mg/ml of G418 (Nacalai Tesque, Inc.). BINDS-30 [MCF7/HER3-knockout (KO) cells] were produced using CRISPR/Cas9 plasmids targeting human HER3 (http://www.med-tohoku-antibody.com/topics/001_paper_cell.htm). Using TruGuide gRNA tool, gRNA of HER3/ERBB3 (NM_001005915) was selected from GeneArt predesigned gRNAs database (Thermo Fisher Scientific, Inc.). gRNA sequence used was GTCCCGTGAGCACAATCTCA(agg), which targeted exon 3 of HER3 (Assay ID: CRISPR764358). Double strand gRNA sequence was subcloned into GeneArt CRISPR Nuclease Vector with OFP Reporter (Thermo Fisher Scientific, Inc.). P3U1, CHO-K1, CHO/PA16-HER3, CHO/HER3-PA, COLO 201, COLO 205, SW1116, DLD-1, MCF7, and BINDS-30 were cultured in Roswell Park Memorial Institute (RPMI)-1640 media (Nacalai Tesque, Inc.). LN229, Caco-2, HCT 116, HCT-15, HT-29, LS 174T, and HCT-8 were cultured in Dulbecco's modified Eagle's medium (DMEM; Nacalai Tesque, Inc.). RPMI-1640 and DMEM were supplemented with 10% heat-inactivated fetal bovine serum (FBS; Thermo Fisher Scientific Inc.), 100 U/ml of penicillin (Nacalai Tesque, Inc.), 100 µg/ml streptomycin (Nacalai Tesque, Inc.), and 0.25 µg/ml amphotericin B (Nacalai Tesque, Inc.), and incubated at 37°C in a humidified atmosphere containing 5% CO2.

Preparation of the purified antibodies

Purified mouse IgG (cat. no. I8765) and mouse IgG2a (cat. no. M7769) were purchased from Sigma-Aldrich; Merck KGaA. An anti-HER3 mAb was purified using Protein G-Sepharose (GE Healthcare BioSciences).

Hybridoma production

Female BALB/c mice (6 weeks old) were purchased from CLEA Japan and kept under specific pathogen-free conditions. All animal experiments were conducted in accordance with the relevant guidelines and regulations in order to minimize animal suffering and distress in the laboratory. The Animal Care and Use Committee of Tohoku University approved all the animal experiments (permit no. 2019NiA-001). Mice were euthanized by cervical dislocation under inhalation anesthesia using 2% of isoflurane for both induction and maintenance, and the death was verified to be respiratory and cardiac arrest.

CBIS method was used as previously reported (54) to develop mAbs against HER3 (Fig. 1). Two eight-week-old BALB/c female mice were intraperitoneally (i.p.) immunized with CHO/PA16-HER3 cells (1×108) along with Imject Alum adjuvant (Thermo Fisher Scientific, Inc.) (Fig. 1A). The procedure included three additional immunizations, followed by a final booster injection administered i.p. two days before the spleen cell harvesting. Spleen cells were then fused with P3U1 cells using PEG1500 (Roche Diagnostics) (Fig. 1B). The developed hybridomas were seeded into 96-well plates, and hybridomas were grown at 37°C for 10 days in RPMI-1640 media with HAT Supplement (50X) (cat. no. 21060017; Thermo Fisher Scientific, Inc.) for selection. Supernatants positive for CHO/HER3-PA and negative for CHO-K1 were selected by flow cytometry (Fig. 1C). After limiting dilution, supernatants positive for LN229 were selected by flow cytometry. Finally, anti-HER3 mAb-producing hybridomas were established (Fig. 1D).

Figure 1.

Production of anti-HER3 mAbs. Procedure of the Cell-Based Immunization and Screening (CBIS) method. (A) CHO/PA16-HER3 cells were immunized into BALB/c mice using an intraperitoneal injection. (B) Spleen cells were fused with P3U1 cells. (C) Culture supernatants were screened using flow cytometry. (D) After limiting dilution and several additional screenings, an anti-HER3 mAb was finally established. mAb, monoclonal antibody.

Western blot analysis

Cell pellets were resuspended in phosphate-buffered saline (PBS; Nacalai Tesque, Inc.) with 1% Triton X-100 (cat. no. 168-11805; FUJIFILM Wako Pure Chemical Corp.) and 50 µg/ml aprotinin (product no. 03346-84; Nacalai Tesque, Inc.). Cell debris was removed by centrifugation at 21,880 × g for 10 min at 4°C. Protein concentration was determined by BCA method. Cell lysates (10 µg) were boiled in sodium dodecyl sulfate (SDS) sample buffer (Nacalai Tesque, Inc.). Proteins were electrophoresed on 5–20% polyacrylamide gels (FUJIFILM Wako Pure Chemical Corp.) and transferred onto polyvinylidene difluoride (PVDF) membranes (Merck KGaA). After blocking with 4% skim milk (Nacalai Tesque, Inc.) at room temperature for 30 min, PVDF membranes were incubated with an anti-HER3 mAb (diluted 1:1,000; clone D22C5) and anti-β-actin mAb (1 µg/ml; clone AC-15; cat. no. A1978; Sigma-Aldrich; Merck KGaA) at room temperature for 30 min, followed by incubation with peroxidase-conjugated anti-rabbit immunoglobulins (diluted 1:1,000; cat. no. P0448; Agilent Technologies Inc.) and peroxidase-conjugated anti-mouse immunoglobulins (diluted 1:1,000; cat. no. P0260; Agilent Technologies Inc.), respectively, at room temperature for 30 min. Blots were developed using ImmunoStar LD (cat. no. 290-69904; FUJIFILM Wako Pure Chemical Corp.) or Pierce™ ECL Plus Western Blotting Substrate (cat. no. 32132; Thermo Fisher Scientific, Inc.) and imaged with a Sayaca-Imager (DRC Co., Ltd.). Qcapture Pro software (DRC Co., Ltd) was used for the densitometry.

Flow cytometry analyses

Cells (2×105 cells/ml) were harvested after brief exposure to 0.25% trypsin in 1 mM ethylenediaminetetraacetic acid (EDTA; Nacalai Tesque, Inc.). After being washed with 0.1% bovine serum albumin (BSA, Nacalai Tesque, Inc.) in PBS, cells were treated with 1 µg/ml of anti-HER3 mAbs, for 30 min at 4°C, and with Alexa Fluor 488-conjugated anti-mouse IgG (1:1,000; cat. no. 4408; Cell Signaling Technology, Inc.). Fluorescence data were collected using a flow cytometer: the EC800 Cell Analyzer (Sony Biotechnology Corp.).

Determination of the binding affinity

Cells (2×105 cells/ml) were suspended in 100 µl of serially diluted anti-HER3 mAb (6 ng/ml-25 µg/ml), followed by the addition of Alexa Fluor 488-conjugated anti-mouse IgG (1:200). Fluorescence data were collected using a flow cytometer: The BD FACSLyric (BD Biosciences). The dissociation constant (KD) was calculated by fitting binding isotherms to built-in one-site binding models in GraphPad Prism 8 (GraphPad Software, Inc.).

ADCC activity of an anti-HER3 mAb

ADCC inducement by HER3 was assayed as follows. Four female five-week-old BALB/c nude mice (mean weight, 15±3 g) were purchased from Charles River Laboratories, Inc. Mice were kept under specific pathogen-free condition on an 11-h light/13-h dark cycle at a temperature of 23±2°C and 55±5% humidity with food and water supplied ad libitum during the experimental periods. After euthanasia by cervical dislocation, spleens were removed aseptically, and single-cell suspensions were obtained by forcing spleen tissues through a sterile cell strainer (product no. 352360; Corning, Inc.) with a syringe. Erythrocytes were lysed with a 10-sec exposure to ice-cold distilled water. The splenocytes were washed with DMEM and resuspended in DMEM with 10% FBS; this preparation was designated as effector cells. The target tumor cells were labeled with 10 µg/ml Calcein-AM (Thermo Fisher Scientific, Inc.) and resuspended in the same medium. The target cells were then transferred to 96-well plates, at 2×104 cells/well, and mixed with effector cells at an effector-to-target ratio of 100:1, along with 100 µg/ml of anti-HER3 antibodies or control mouse IgG2a. After a 4.5-h incubation at 37°C, Calcein release into the supernatant was measured for each well. Fluorescence intensity was assessed using a microplate reader (Power Scan HT; BioTek Instruments, Inc.) with an excitation wavelength of 485 nm and an emission wavelength of 538 nm. Cytolytic activity was measured as a percentage of lysis and calculated using the equation: Percentage of lysis (%) = (E-S)/(M-S) ×100, where E is the fluorescence measured in combined cultures of target and effector cells, S is the spontaneous fluorescence of the target cells, and M is the maximum fluorescence measured after lysis of all cells with buffer containing 0.5% Triton X-100, 10 mM Tris-HCl (pH 7.4), and 10 mM EDTA. Animal studies for ADCC were approved by the Institutional Committee for experiments of the Institute of Microbial Chemistry (permit no. 2020-024).

CDC activity of an anti-HER3 mAb

CDC inducement by HER3 was assayed as follows. Target cells were labeled with 10 µg/ml Calcein-AM (Thermo Fisher Scientific, Inc.), resuspended in medium and plated in 96-well plates, at 2×104 cells/well, with 15% rabbit complement (Low-Tox-M rabbit complement; Cedarlane Laboratories), 100 µg/ml of anti-HER3 antibodies, or control IgG (mouse IgG2a) added to each well. After 4.5 h of incubation at 37°C, Calcein release into the supernatant was measured for each well. Fluorescence intensity was calculated as described in the ADCC section above.

Antitumor activity of an anti-HER3 mAb in xenografts of colon cancers

Sixteen five-week-old female BALB/c nude mice (mean weight, 15±3 g) were purchased from Charles River Laboratories, Inc. All animal experiments were performed in accordance with institutional guidelines and regulations to minimize animal suffering and distress in the laboratory. The Institutional Committee for experiments of the Institute of Microbial Chemistry (permit no. 2020-024) approved the animal studies for antitumor activity here described. Mice were maintained in a pathogen-free environment on an 11-h light/13-h dark cycle at a temperature of 23±2°C and 55±5% humidity with food and water supplied ad libitum throughout the experiments. Mice were monitored for health and weight every three or five days. Experiments on mice were conducted in four weeks. Weight loss exceeding 25% or tumor volume exceeding 3,000 mm3 were identified as humane endpoints for euthanasia. At humane and experimental endpoints, mice were euthanized by cervical dislocation, and death was verified by validating respiratory and cardiac arrest.

After a one-week acclimation period, these mice were used in experiments at six weeks of age (mean weight, 16±2 g). Caco-2 cells (0.3 ml of 1.33×108 cells/ml in DMEM) were mixed with 0.5 ml BD Matrigel Matrix Growth Factor Reduced (BD Biosciences), and 100 µl of this suspension (5×106 cells) was injected subcutaneously into the left flank of each animal. On the eighth day post-inoculation, 16 mice were divided into two groups (n=8 in each group) with equal mean tumor volume: An anti-HER3 mAb group or a control mouse IgG group. Then, 100 µg of an anti-HER3 mAb or control mouse IgG in 100 µl PBS was injected i.p. Additional antibody inoculations were performed on days 15 and 23. Twenty-six days after cell implantation, all mice were euthanized by cervical dislocation, and tumor diameters and volumes were measured and recorded.

Statistical analyses

All data are expressed as mean ± standard error of the mean (SEM). Statistical analysis was conducted with ANOVA and Tukey's multiple comparisons tests for ADCC and CDC, ANOVA and Sidak's multiple comparisons tests for tumor volume and mouse weight, and Welch's t-test for tumor weight. All calculations were performed with GraphPad Prism 8 (GraphPad Software, Inc.). A P-value of <0.05 was considered to indicate a statistically significant difference.

Results

Development of anti-HER3 mAbs

We employed the CBIS method to develop anti-HER3 mAbs using CHO/PA16-HER3 cells both for the immunization and flow cytometry screening (Fig. 1). The developed hybridomas were seeded into 96-well plates and cultivated for 10 days. Supernatants positive for CHO/HER3-PA and negative for CHO-K1 were selected by flow cytometry. After limiting dilution and several additional screenings, an anti-HER3 mAb, H3Mab-17 (mouse IgG2a, kappa), was finally established.

Confirmation of HER3 expression by western blot analysis

We established CHO/mock (Ble), CHO/PA16-HER3, CHO/mock (Neo), and CHO/HER3-PA, and investigated whether HER3 was overexpressed in those cell lines. As shown in Fig. 2A, overexpression of HER3 in CHO/PA16-HER3 and CHO/HER3-PA was confirmed by western blot analysis using an anti-HER3 mAb (clone D22C5). Endogenous HER3 expression in MCF7 cells was also detected by an anti-HER3 mAb. In contrast, knockout of endogenous HER3 in BINDS-30 (MCF7/HER3-KO) was confirmed by western blot analysis using an anti-HER3 mAb.

Figure 2.

Characterization of H3Mab-17. (A) Confirmation of HER3 expression by western blot analysis. Cell lysates were electrophoresed and transferred onto PVDF membranes. After blocking, PVDF membranes were incubated with an anti-HER3 mAb (clone D22C5) or anti-β-actin (clone AC-15), followed by incubation with peroxidase-conjugated anti-rabbit immunoglobulins or peroxidase-conjugated anti-mouse immunoglobulins. Blots were developed using ImmunoStar LD or ECL Plus Western Blotting Substrate and imaged with a Sayaca-Imager. (B) Flow cytometry analysis. CHO-K1, CHO/HER3-PA, MCF7, and BINDS-30 cells were treated with 1 µg/ml of H3Mab-17, followed by treatment with Alexa Fluor 488-conjugated anti-mouse IgG. Black line, negative control. (C) Flow cytometry analysis. Colon cancer cell lines, such as Caco-2, HCT 116, HCT-15, HT-29, LS 174T, COLO 201, COLO 205, HCT-8, SW1116, and DLD-1 cells were treated with 1 µg/ml of H3Mab-17, followed by treatment with Alexa Fluor 488-conjugated anti-mouse IgG. Black line, negative control. mAb, monoclonal antibody.

Flow cytometry analyses of H3Mab-17

We performed flow cytometry using H3Mab-17 against CHO-K1, CHO/HER3-PA, MCF7, and BINDS-30 (MCF7/HER3-KO). H3Mab-17 recognized the CHO/HER3-PA cells, but not the parental CHO-K1 cells (Fig. 2B). H3Mab-17 also recognized the endogenous HER3 in MCF7 breast cancer cells (Fig. 2B). The reaction of H3Mab-17 to BINDS-30 was lost after the knockout of HER3 in MCF7 cells (Fig. 2B), indicating the specificity of H3Mab-17 for HER3.

Next, we investigated whether H3Mab-17 reacts with colon cancer cell lines. As shown in Fig. 2C, H3Mab-17 reacted with 10 colon cancer cell lines, Caco-2, HCT 116, HCT-15, HT-29, LS 174T, COLO 201, COLO 205, HCT-8, SW1116, and DLD-1. Among them, Caco-2 was known to be useful for the mouse xenograft model (55). Therefore, we used Caco-2 cells for the ADCC/CDC assay or in vivo xenograft models.

Determination of the binding affinity of H3Mab-17

A kinetic analysis of the interactions of H3Mab-17 with CHO/HER3-PA and Caco-2 cells was then conducted using flow cytometry. The KD for H3Mab-17 in CHO/HER3-PA and Caco-2 cells were 3.0×10−9 and 1.5×10−9 M, respectively (Fig. 3), indicating high binding affinity of H3Mab-17 against HER3-expressing cells.

Figure 3.

Determination of the binding affinity of H3Mab-17. CHO/HER3-PA and Caco-2 were suspended in 100 µl of serially diluted H3Mab-17 (0.006–25 µg/ml) prior to Alexa Fluor 488-conjugated anti-mouse IgG addition. Fluorescence data were collected using a BD FACSLyric. KD, dissociation constant.

ADCC and CDC activities of H3Mab-17 in colon cancer cell lines

We then examined whether H3Mab-17 (mouse IgG2a) induced ADCC and CDC activity in HER3-expressing Caco-2 colon cancer cell lines. H3Mab-17 exhibited higher ADCC (14.8% cytotoxicity) in Caco-2 cells than that of control mouse IgG2a (5.2% cytotoxicity; P<0.05) or control PBS (3.2% cytotoxicity; P<0.05) treatment (Fig. 4A). H3Mab-17 was also associated with a more robust CDC activity (30.4% cytotoxicity) in Caco-2 cells than the control mouse IgG2a (7.7% cytotoxicity; P<0.05) or the control with PBS treatment (8.8% cytotoxicity; P<0.05) (Fig. 4B). These favorable ADCC/CDC activities indicated that H3Mab-17 may induce strong antitumor activity against colon cancer cells in vivo as well as in vitro.

Figure 4.

Evaluation of ADCC and CDC activities by H3Mab-17. (A) ADCC activities by H3Mab-17, control mouse IgG2a, and control PBS in Caco-2 cells. (B) CDC activities by H3Mab-17, control mouse IgG2a, and control PBS in Caco-2 cells. Values are mean ± SEM. Asterisk indicates statistical significance (*P<0.05; n.s., not significant, ANOVA and Tukey-Kramer's test). ADCC, antibody-dependent cellular cytotoxicity; CDC, complement-dependent cytotoxicity.

Antitumor effect of H3Mab-17 in mouse xenografts of colon cancer cells

Tumor formation of 16 Caco-2-bearing mice was observed on day eight. Then, these 16 Caco-2-bearing mice were divided into an H3Mab-17-treated group and a control group. On days 8, 15 and 23 after Caco-2 cell injections into the mice, H3Mab-17 (100 µg) or control mouse IgG (100 µg) were injected i.p. in the Caco-2 ×enograft model mice. The tumor volume was measured on days 8, 11, 15, 18, 23 and 26 after the Caco-2 cell injection. H3Mab-17-treated mice exhibited significantly less tumor growth on day 18 (P<0.01), day 23 (P<0.01), and day 26 (P<0.01), compared with the IgG-treated control mice (Fig. 5A). The reduction in the volume of the tumors by H3Mab-17 treatment was of 54% on day 26. Tumors from the H3Mab-17-treated mice weighed significantly less than tumors from the IgG-treated control mice (59% reduction, P<0.01; Fig. 5B). Resected tumors on day 26 are presented in Fig. 5C. The total body weights did not significantly differ between the treatment and control groups (Fig. 6A and B). These results indicated that H3Mab-17 reduced the growth of Caco-2 ×enografts, without eliminating them completely.

Figure 5.

Evaluation of antitumor activity of H3Mab-17 in Caco-2 ×enografts. (A) Caco-2 cells (5×106 cells) were injected subcutaneously into the left flank. After day 8, 100 µg of H3Mab-17 and control mouse IgG in 100 µl PBS were injected i.p. into the treated and control mice, respectively. Additional antibodies were then injected on days 15 and 23. Tumor volume was measured on days 8, 11, 15, 18, 23 and 26. Values are mean ± SEM. Asterisk indicates statistical significance (**P<0.01; n.s., not significant, ANOVA and Sidak's multiple comparisons test). ○, control; ∆, H3Mab-17. (B) Tumors of Caco-2 ×enografts were resected from H3Mab-17 and control mouse IgG groups. Tumor weight on day 26 was measured from excised xenografts. Values are mean ± SEM. Asterisk indicates statistical significance (**P<0.01, Welch's t-test). ○, control; ∆, H3Mab-17. (C) Resected tumors of Caco-2 ×enografts from H3Mab-17 and control mouse IgG groups on day 26. The tumor in the square dotted region was the largest tumor in this experiment. The vertical and horizontal lengths for Caco-2 cells were 1.6 and 1.3 cm, respectively (estimated tumor volume, 1,352 mm3, tumor weight, 661 mg). Scale bar, 1 cm.

Figure 6.

Body weights of mice implanted with Caco-2 ×enografts. Caco-2 cells (5×106 cells) were injected subcutaneously into the left flank. After day 8, 100 µg of H3Mab-17 and control mouse IgG in 100 µl PBS were injected i.p. into treated and control mice, respectively. Additional antibodies were then injected on days 15 and 23. (A) Body weights of the mice with the Caco-2 ×enografts of H3Mab-17 group and control group were recorded on days 8, 11, 15, 18, 23 and 26. Values are mean ± SEM. n.s. indicates no statistical significance (ANOVA and Sidak's multiple comparisons test). (B) Body appearance of mice with the Caco-2 ×enografts of H3Mab-17 and control mouse IgG groups on day 26. Scale bar, 1 cm.

Discussion

Many commercially available anti-HER3 mAbs have been developed using recombinant HER3 protein, peptide or cDNA as an immunogen. Seribantumab was developed by phage display (56,57) and lumretuzumab was developed using recombinant HER3 extracellular domain as an immunogen (46). In this study, we succeeded in the development of an anti-HER3 mAb using the CBIS method, which used HER3-expressed cells for both immunization and screening. The CBIS method can help us effectively develop mAbs that are useful in flow cytometry. We recently succeeded in developing numerous useful mAbs that target membrane proteins, including podoplanin (58–61), CD20 (62), CD44 (63), CD133 (54), and TROP2 (64,65). Importantly, these mAbs are very useful for various experiments, including not only flow cytometry, but also western blot analysis and immunohistochemistry. Furthermore, those mAbs possess ADCC/CDC activities and antitumor activities (61). Using the CBIS method, proteins for immunogen expressed on cells maintain its native conformation and glycosylation pattern. Previously, we successfully established a cancer-specific mAb (CasMab) against podoplanin, which recognizes the cancer cell-specific glycosylation of podoplanin (66). Therefore, we may develop CasMab against HER3 using the CBIS method in the future. The CBIS method is advantageous for the development for specific and sensitive mAbs for antibody therapy.

New highly accurate therapeutic options are possible to treat most solid tumors. In the case of colorectal cancer, HER3 overexpression is found in ~17-75%, although the definition of its cutoff signals for HER3 expression are different in each immunohistochemical study (67). It has been reported that the incidence of HER3 overexpression in metastatic colorectal cancer is much higher than that of HER2 (68). In this study, we developed an anti-HER3 mAb, H3Mab-17, which specifically reacted with endogenous HER3 in colorectal carcinoma cell lines in flow cytometry. The KD for H3Mab-17 in CHO/HER3-PA and Caco-2 cells were determined to be 3.0×10−9 and 1.5×10−9 M, respectively, suggesting high binding affinity of H3Mab-17 for HER3. In vitro experiments revealed strong ADCC and CDC inducement against Caco-2 cells by H3Mab-17. In vivo experiments on Caco-2 ×enografts revealed that the treatment with H3Mab-17 significantly reduced the tumor growth, compared with the control mouse IgG. Based on these findings, H3Mab-17 may be useful in therapeutic approach for patients with colorectal cancer.

Although H3Mab-17 recognizes both overexpressed and endogenous HER3 by flow cytometric analyses, it is not applicable to western blot and immunohistochemical analyses (data not shown). H3Mab-17 did not recognize denatured HER3, such as SDS-treated and formalin-fixed HER3 probably because it might recognize the three-dimensional structure of HER3. Since the antitumor activity mechanism of H3Mab-17 has not been clarified, we need to identify the epitope of H3Mab-17 and investigate the inhibitory activity of HER3-neureglin interaction of H3Mab-17. Furthermore, HER3-ADC and HER3-chimeric antigen receptor (CAR)-T should be developed in future research.